base_model: /root/autodl-tmp/SillyTilly/mistralai_Mistral-Nemo-Base-2407 | |
model_type: AutoModelForCausalLM | |
tokenizer_type: AutoTokenizer | |
plugins: | |
- axolotl.integrations.liger.LigerPlugin | |
liger_rope: true | |
liger_rms_norm: true | |
liger_swiglu: true | |
liger_fused_linear_cross_entropy: true | |
load_in_8bit: false | |
load_in_4bit: false | |
strict: false | |
datasets: | |
- path: taozi555/novel_text | |
name: default | |
type: completion | |
- path: ToastyPigeon/SpringDragon | |
type: completion | |
- path: allura-org/sugarquill-10k | |
type: completion | |
- path: allura-org/fujin-cleaned-stage-2 | |
type: completion | |
#chat_template: chatml | |
shuffle_merged_datasets: true | |
#default_system_message: "You are an assistant that responds to the user." | |
dataset_prepared_path: /root/autodl-tmp/magnum-12b-data | |
val_set_size: 0.0 | |
output_dir: /root/autodl-tmp/out | |
sequence_len: 32768 | |
sample_packing: true | |
pad_to_sequence_len: true | |
adapter: | |
lora_model_dir: | |
lora_r: | |
lora_alpha: | |
lora_dropout: | |
lora_target_linear: | |
lora_fan_in_fan_out: | |
wandb_project: 12b-magnum-fft | |
wandb_entity: | |
wandb_watch: | |
wandb_name: v4-r2-attempt-01 | |
wandb_log_model: | |
gradient_accumulation_steps: 2 | |
micro_batch_size: 1 | |
num_epochs: 2 | |
optimizer: adamw_torch | |
lr_scheduler: cosine | |
learning_rate: 0.00003 | |
train_on_inputs: false | |
group_by_length: false | |
bf16: auto | |
fp16: | |
tf32: false | |
gradient_checkpointing: true | |
early_stopping_patience: | |
resume_from_checkpoint: | |
local_rank: | |
logging_steps: 1 | |
xformers_attention: | |
flash_attention: true | |
warmup_steps: 40 | |
evals_per_epoch: | |
eval_table_size: | |
eval_max_new_tokens: | |
saves_per_epoch: 2 | |
debug: | |
deepspeed: /root/autodl-tmp/zero2.json | |
weight_decay: 0.1 | |
fsdp: | |
fsdp_config: | |
special_tokens: | |
pad_token: <pad> | |